Hyperspectral Target Detection Based on Balanced Distance Sub Spectra Selection

Accurate and fast target detection is one of the key problems in hyperspectral image applications.Band selection is an essential step to improve the utilization efficiency of hyperspectral data.The current hyperspectral band selection methods don't consider task correlation,which affects the effectiveness of band selection results in actual target detection tasks.A new spectra equal interval extraction method was proposed based on constructing the spectra interval difference equalization calculation model.Experiments on hyperspectral remote sensing image dataset have confirmed the superiority of the new method.The results show that the proposed method can achieve better results in terms of computational time and accuracy compared with other band selection methods,and can efficiently achieve target detection of hyperspectral images.